Although the pathological correlates responsible for HIV-induced neurological disease are poorly understood, recent evidence suggests that alterations in the blood-brain barrier (BBB) contribute to the progression of HIV-induced dementia. Influx of serum components into the central nervous system could compromise neuronal function by altering the micro-environment. HIV-infected individuals have systemic abnormalities, such as immune activation and elevated cytokine levels, that could alter the BBB and produce neuronal dysfunction and eventual neuronal death. The overall objective of this study is to determine the contribution of altered BBB to the progression of neurological diseases associated with HIV infection. Specific Aim 1 will define the extent of BBB damage during HIV infection and correlate this damage with the clinical and immunological status of the patient. If alterations in the BBB contribute to neurologic disease in HIV-infected individuals, such disease may be prevented or delayed by therapies that prevent BBB damage. Therefore, the studies in Specific aims 2-4 will define changes in endothelial cells and perivascular macrophages that contribute to BBB dysfunction, whereas Specific Aim 5 will characterize alterations in neurons. Preliminary studies indicate that production of nitric oxide (NO) is upregulated in endothelial cells of HIV-infected brains. Endothelial NO is a vasodilator and toxin that can alter the permeability of the BBB. Because NO is difficult to measure, NO synthesizing cells will be identified by localizing the NO metabolizing enzyme, nitric oxide synthase (NOS). At least four different forms of NOS are present in brain. Specific Aim 2 will determine the distribution and quantity of these different NOS molecules in HIV-infected brains. We will determine if increases in NOS expression correlate with alterations in the BBB or with dementia. Specific Aim 3 of this proposal will characterize the expression of adhesion molecules by endothelial cells in HIV-infected brains. It is anticipated that endothelial cells will alter the expression of genes that are known to be inducible or to respond to pathological conditions. We will determine if the expression of ICAM, VCAM, and ELAM on endothelial cells is related to breakdown of the BBB or dementia. These adhesion molecules can bind monocytes and they may facilitate the entry of perivascular macrophages into HIV-infected brain. Specific Aim 4 will focus on the characterization of perivascular macrophages in HIV-infected brain. We will determine whether the numbers of these cells are increased in HIV infection and whether their increase correlates with alterations in the BBB or with dementia. In addition, we will determine what percentage of perivascular macrophages express detectible levels of HIV protein and cytokines such as TNF-alpha and IL-1beta. It is anticipated that alterations in endothelial cells will precede neurological dysfunction and detectible changes in the BBB. Specific Aim 5 will correlate the development of neuronal pathology with changes in endothelial cells, BBB breakdown, and development of dementia.